{
  "best_metric": 0.23757965862751007,
  "best_model_checkpoint": "qa-model-finetune/checkpoint-117",
  "epoch": 8.26,
  "eval_steps": 500,
  "global_step": 117,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.1156,
      "step": 1
    },
    {
      "epoch": 0.04,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.225,
      "step": 2
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.2e-05,
      "loss": 1.0757,
      "step": 3
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.8098,
      "step": 4
    },
    {
      "epoch": 0.1,
      "learning_rate": 2e-05,
      "loss": 1.2561,
      "step": 5
    },
    {
      "epoch": 0.12,
      "learning_rate": 2.4e-05,
      "loss": 1.2355,
      "step": 6
    },
    {
      "epoch": 0.14,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 1.0707,
      "step": 7
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.3054,
      "step": 8
    },
    {
      "epoch": 0.18,
      "learning_rate": 3.6e-05,
      "loss": 1.2546,
      "step": 9
    },
    {
      "epoch": 0.2,
      "learning_rate": 4e-05,
      "loss": 1.2499,
      "step": 10
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 1.0749,
      "step": 11
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.8e-05,
      "loss": 1.1383,
      "step": 12
    },
    {
      "epoch": 0.26,
      "learning_rate": 5.2000000000000004e-05,
      "loss": 1.15,
      "step": 13
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.841742217540741,
      "eval_runtime": 24.118,
      "eval_samples_per_second": 0.788,
      "eval_steps_per_second": 0.207,
      "step": 13
    },
    {
      "epoch": 1.02,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 0.9705,
      "step": 14
    },
    {
      "epoch": 1.04,
      "learning_rate": 6e-05,
      "loss": 1.2398,
      "step": 15
    },
    {
      "epoch": 1.06,
      "learning_rate": 6.400000000000001e-05,
      "loss": 0.9997,
      "step": 16
    },
    {
      "epoch": 1.08,
      "learning_rate": 6.800000000000001e-05,
      "loss": 1.2356,
      "step": 17
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.2e-05,
      "loss": 0.8312,
      "step": 18
    },
    {
      "epoch": 1.12,
      "learning_rate": 7.6e-05,
      "loss": 1.0169,
      "step": 19
    },
    {
      "epoch": 1.14,
      "learning_rate": 8e-05,
      "loss": 0.8709,
      "step": 20
    },
    {
      "epoch": 1.16,
      "learning_rate": 8.4e-05,
      "loss": 1.121,
      "step": 21
    },
    {
      "epoch": 1.18,
      "learning_rate": 8.800000000000001e-05,
      "loss": 0.9481,
      "step": 22
    },
    {
      "epoch": 1.2,
      "learning_rate": 9.200000000000001e-05,
      "loss": 0.8888,
      "step": 23
    },
    {
      "epoch": 1.22,
      "learning_rate": 9.6e-05,
      "loss": 1.0233,
      "step": 24
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001,
      "loss": 0.8382,
      "step": 25
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00010400000000000001,
      "loss": 0.8734,
      "step": 26
    },
    {
      "epoch": 1.26,
      "eval_loss": 0.7118059992790222,
      "eval_runtime": 24.0715,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.208,
      "step": 26
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.00010800000000000001,
      "loss": 0.9912,
      "step": 27
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.00011200000000000001,
      "loss": 1.0997,
      "step": 28
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.000116,
      "loss": 0.7957,
      "step": 29
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.00012,
      "loss": 0.9109,
      "step": 30
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.000124,
      "loss": 0.9713,
      "step": 31
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.00012800000000000002,
      "loss": 0.7388,
      "step": 32
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.000132,
      "loss": 0.481,
      "step": 33
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.00013600000000000003,
      "loss": 0.9344,
      "step": 34
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.00014,
      "loss": 0.8717,
      "step": 35
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.000144,
      "loss": 0.7509,
      "step": 36
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.000148,
      "loss": 0.8116,
      "step": 37
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.000152,
      "loss": 0.7251,
      "step": 38
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.00015600000000000002,
      "loss": 0.7168,
      "step": 39
    },
    {
      "epoch": 2.26,
      "eval_loss": 0.5603676438331604,
      "eval_runtime": 23.997,
      "eval_samples_per_second": 0.792,
      "eval_steps_per_second": 0.208,
      "step": 39
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.00016,
      "loss": 0.7851,
      "step": 40
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.000164,
      "loss": 0.7412,
      "step": 41
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.000168,
      "loss": 0.674,
      "step": 42
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.000172,
      "loss": 0.7634,
      "step": 43
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.00017600000000000002,
      "loss": 0.6933,
      "step": 44
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.00018,
      "loss": 0.5634,
      "step": 45
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.00018400000000000003,
      "loss": 0.6571,
      "step": 46
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.000188,
      "loss": 0.5984,
      "step": 47
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.000192,
      "loss": 0.674,
      "step": 48
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.000196,
      "loss": 0.6171,
      "step": 49
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0002,
      "loss": 0.6129,
      "step": 50
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.00019955555555555558,
      "loss": 0.5328,
      "step": 51
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.00019911111111111111,
      "loss": 0.479,
      "step": 52
    },
    {
      "epoch": 3.26,
      "eval_loss": 0.41886183619499207,
      "eval_runtime": 23.9197,
      "eval_samples_per_second": 0.794,
      "eval_steps_per_second": 0.209,
      "step": 52
    },
    {
      "epoch": 4.02,
      "learning_rate": 0.00019866666666666668,
      "loss": 0.5089,
      "step": 53
    },
    {
      "epoch": 4.04,
      "learning_rate": 0.00019822222222222225,
      "loss": 0.5049,
      "step": 54
    },
    {
      "epoch": 4.06,
      "learning_rate": 0.00019777777777777778,
      "loss": 0.5407,
      "step": 55
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.00019733333333333335,
      "loss": 0.3821,
      "step": 56
    },
    {
      "epoch": 4.1,
      "learning_rate": 0.0001968888888888889,
      "loss": 0.484,
      "step": 57
    },
    {
      "epoch": 4.12,
      "learning_rate": 0.00019644444444444445,
      "loss": 0.5705,
      "step": 58
    },
    {
      "epoch": 4.14,
      "learning_rate": 0.000196,
      "loss": 0.5301,
      "step": 59
    },
    {
      "epoch": 4.16,
      "learning_rate": 0.00019555555555555556,
      "loss": 0.6168,
      "step": 60
    },
    {
      "epoch": 4.18,
      "learning_rate": 0.0001951111111111111,
      "loss": 0.4745,
      "step": 61
    },
    {
      "epoch": 4.2,
      "learning_rate": 0.0001946666666666667,
      "loss": 0.6383,
      "step": 62
    },
    {
      "epoch": 4.22,
      "learning_rate": 0.00019422222222222223,
      "loss": 0.3682,
      "step": 63
    },
    {
      "epoch": 4.24,
      "learning_rate": 0.0001937777777777778,
      "loss": 0.5654,
      "step": 64
    },
    {
      "epoch": 4.26,
      "learning_rate": 0.00019333333333333333,
      "loss": 0.451,
      "step": 65
    },
    {
      "epoch": 4.26,
      "eval_loss": 0.3399796783924103,
      "eval_runtime": 24.1635,
      "eval_samples_per_second": 0.786,
      "eval_steps_per_second": 0.207,
      "step": 65
    },
    {
      "epoch": 5.02,
      "learning_rate": 0.0001928888888888889,
      "loss": 0.4663,
      "step": 66
    },
    {
      "epoch": 5.04,
      "learning_rate": 0.00019244444444444444,
      "loss": 0.4582,
      "step": 67
    },
    {
      "epoch": 5.06,
      "learning_rate": 0.000192,
      "loss": 0.3941,
      "step": 68
    },
    {
      "epoch": 5.08,
      "learning_rate": 0.00019155555555555554,
      "loss": 0.401,
      "step": 69
    },
    {
      "epoch": 5.1,
      "learning_rate": 0.00019111111111111114,
      "loss": 0.5281,
      "step": 70
    },
    {
      "epoch": 5.12,
      "learning_rate": 0.00019066666666666668,
      "loss": 0.4658,
      "step": 71
    },
    {
      "epoch": 5.14,
      "learning_rate": 0.00019022222222222224,
      "loss": 0.4215,
      "step": 72
    },
    {
      "epoch": 5.16,
      "learning_rate": 0.00018977777777777778,
      "loss": 0.33,
      "step": 73
    },
    {
      "epoch": 5.18,
      "learning_rate": 0.00018933333333333335,
      "loss": 0.5385,
      "step": 74
    },
    {
      "epoch": 5.2,
      "learning_rate": 0.00018888888888888888,
      "loss": 0.3539,
      "step": 75
    },
    {
      "epoch": 5.22,
      "learning_rate": 0.00018844444444444445,
      "loss": 0.4084,
      "step": 76
    },
    {
      "epoch": 5.24,
      "learning_rate": 0.000188,
      "loss": 0.2773,
      "step": 77
    },
    {
      "epoch": 5.26,
      "learning_rate": 0.00018755555555555558,
      "loss": 0.3543,
      "step": 78
    },
    {
      "epoch": 5.26,
      "eval_loss": 0.29384246468544006,
      "eval_runtime": 24.1775,
      "eval_samples_per_second": 0.786,
      "eval_steps_per_second": 0.207,
      "step": 78
    },
    {
      "epoch": 6.02,
      "learning_rate": 0.00018711111111111112,
      "loss": 0.4595,
      "step": 79
    },
    {
      "epoch": 6.04,
      "learning_rate": 0.0001866666666666667,
      "loss": 0.4364,
      "step": 80
    },
    {
      "epoch": 6.06,
      "learning_rate": 0.00018622222222222223,
      "loss": 0.2679,
      "step": 81
    },
    {
      "epoch": 6.08,
      "learning_rate": 0.0001857777777777778,
      "loss": 0.3882,
      "step": 82
    },
    {
      "epoch": 6.1,
      "learning_rate": 0.00018533333333333333,
      "loss": 0.3004,
      "step": 83
    },
    {
      "epoch": 6.12,
      "learning_rate": 0.0001848888888888889,
      "loss": 0.3132,
      "step": 84
    },
    {
      "epoch": 6.14,
      "learning_rate": 0.00018444444444444446,
      "loss": 0.4519,
      "step": 85
    },
    {
      "epoch": 6.16,
      "learning_rate": 0.00018400000000000003,
      "loss": 0.2922,
      "step": 86
    },
    {
      "epoch": 6.18,
      "learning_rate": 0.00018355555555555557,
      "loss": 0.3561,
      "step": 87
    },
    {
      "epoch": 6.2,
      "learning_rate": 0.00018311111111111113,
      "loss": 0.3193,
      "step": 88
    },
    {
      "epoch": 6.22,
      "learning_rate": 0.00018266666666666667,
      "loss": 0.3153,
      "step": 89
    },
    {
      "epoch": 6.24,
      "learning_rate": 0.00018222222222222224,
      "loss": 0.2202,
      "step": 90
    },
    {
      "epoch": 6.26,
      "learning_rate": 0.00018177777777777778,
      "loss": 0.2222,
      "step": 91
    },
    {
      "epoch": 6.26,
      "eval_loss": 0.2629795968532562,
      "eval_runtime": 24.1398,
      "eval_samples_per_second": 0.787,
      "eval_steps_per_second": 0.207,
      "step": 91
    },
    {
      "epoch": 7.02,
      "learning_rate": 0.00018133333333333334,
      "loss": 0.1878,
      "step": 92
    },
    {
      "epoch": 7.04,
      "learning_rate": 0.0001808888888888889,
      "loss": 0.2644,
      "step": 93
    },
    {
      "epoch": 7.06,
      "learning_rate": 0.00018044444444444447,
      "loss": 0.2501,
      "step": 94
    },
    {
      "epoch": 7.08,
      "learning_rate": 0.00018,
      "loss": 0.2801,
      "step": 95
    },
    {
      "epoch": 7.1,
      "learning_rate": 0.00017955555555555558,
      "loss": 0.3479,
      "step": 96
    },
    {
      "epoch": 7.12,
      "learning_rate": 0.00017911111111111112,
      "loss": 0.3556,
      "step": 97
    },
    {
      "epoch": 7.14,
      "learning_rate": 0.00017866666666666668,
      "loss": 0.2206,
      "step": 98
    },
    {
      "epoch": 7.16,
      "learning_rate": 0.00017822222222222222,
      "loss": 0.2939,
      "step": 99
    },
    {
      "epoch": 7.18,
      "learning_rate": 0.00017777777777777779,
      "loss": 0.2029,
      "step": 100
    },
    {
      "epoch": 7.2,
      "learning_rate": 0.00017733333333333335,
      "loss": 0.2979,
      "step": 101
    },
    {
      "epoch": 7.22,
      "learning_rate": 0.0001768888888888889,
      "loss": 0.2793,
      "step": 102
    },
    {
      "epoch": 7.24,
      "learning_rate": 0.00017644444444444446,
      "loss": 0.2809,
      "step": 103
    },
    {
      "epoch": 7.26,
      "learning_rate": 0.00017600000000000002,
      "loss": 0.2748,
      "step": 104
    },
    {
      "epoch": 7.26,
      "eval_loss": 0.25376519560813904,
      "eval_runtime": 24.0872,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.208,
      "step": 104
    },
    {
      "epoch": 8.02,
      "learning_rate": 0.00017555555555555556,
      "loss": 0.116,
      "step": 105
    },
    {
      "epoch": 8.04,
      "learning_rate": 0.00017511111111111113,
      "loss": 0.2272,
      "step": 106
    },
    {
      "epoch": 8.06,
      "learning_rate": 0.00017466666666666667,
      "loss": 0.1273,
      "step": 107
    },
    {
      "epoch": 8.08,
      "learning_rate": 0.00017422222222222223,
      "loss": 0.1605,
      "step": 108
    },
    {
      "epoch": 8.1,
      "learning_rate": 0.0001737777777777778,
      "loss": 0.2263,
      "step": 109
    },
    {
      "epoch": 8.12,
      "learning_rate": 0.00017333333333333334,
      "loss": 0.2452,
      "step": 110
    },
    {
      "epoch": 8.14,
      "learning_rate": 0.0001728888888888889,
      "loss": 0.2009,
      "step": 111
    },
    {
      "epoch": 8.16,
      "learning_rate": 0.00017244444444444444,
      "loss": 0.2372,
      "step": 112
    },
    {
      "epoch": 8.18,
      "learning_rate": 0.000172,
      "loss": 0.2736,
      "step": 113
    },
    {
      "epoch": 8.2,
      "learning_rate": 0.00017155555555555555,
      "loss": 0.3009,
      "step": 114
    },
    {
      "epoch": 8.22,
      "learning_rate": 0.0001711111111111111,
      "loss": 0.2619,
      "step": 115
    },
    {
      "epoch": 8.24,
      "learning_rate": 0.00017066666666666668,
      "loss": 0.2698,
      "step": 116
    },
    {
      "epoch": 8.26,
      "learning_rate": 0.00017022222222222224,
      "loss": 0.1716,
      "step": 117
    },
    {
      "epoch": 8.26,
      "eval_loss": 0.23757965862751007,
      "eval_runtime": 24.2556,
      "eval_samples_per_second": 0.783,
      "eval_steps_per_second": 0.206,
      "step": 117
    }
  ],
  "logging_steps": 1,
  "max_steps": 500,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.8291357646848e+16,
  "trial_name": null,
  "trial_params": null
}